[MapServer-users] UVRASTER reprojection

Tue Oct 14 05:28:49 PDT 2025

Hi Seth,

Thanks for taking a look. I've had a look at some of the existing test
cases and managed to make some tests which show the issue more clearly. I
can open an issue and link to a branch with these updated tests, I'll try
to do this later today or tomorrow.

I've written an algorithm in python which we will use in the mean time to
pre-calculate the corrected angles. I'm more than happy to share that code
if someone would like to try to port it to mapserver. I can also take a
look if I have time, but have to admit my C++ isn't very up to scratch. To
avoid taking a lot of time calculating on the fly it would also be smart to
cache the point-wise rotation angle used to convert between different
projections, and I'm unsure how this would be handled in mapserver.

Thanks,
Alistair

On Sat, 11 Oct 2025 at 12:01, Seth G <sethg at geographika.co.uk> wrote:

> Hi Alistair,
>
> The Mapfile and requests look fine to me, so this is likely an oversight
> in the implementation.
> It would be good to get this logged as an issue, ideally with a
> recreatable test case at https://github.com/MapServer/MapServer/issues.
>
> Seth
>
> --
> web:https://geographika.net & https://mapserverstudio.net
> mastodon: @geographika at mastodon.social
>
> On Mon, Oct 6, 2025, at 12:16 PM, Alistair Everett wrote:
>
> Hi Seth,
>
> Thanks for your reply, we're just reprojecting the data. I'll take a
> closer look at the test case too, since it seems like that should be
> reprojecting from stereographic to geographic. I should also be able to
> pull together some test data in the next few days too, I can open an issue
> with it when I've put it together.
>
> Here are a couple of example requests:
>
> One in EPSG:4326
>
> localhost:8000/wms?REQUEST=GetMap&SERVICE=WMS&VERSION=1.3.0&FORMAT=image%2Fpng&STYLES=wind_barb&TRANSPARENT=TRUE&TIME=2025-10-06T07%3A00%3A00Z&DIM_PRESSURE=50.0&DIM_REFERENCE_TIME=2025-10-06T03%3A00%3A00Z&LAYERS=wind_pl_vector_meps_det_vdiv_2_5km_calculations&WIDTH=1304&HEIGHT=1504&CRS=EPSG%3A3857&BBOX=-2460529.2703191666%2C4596419.518351211%2C6809846.057963322%2C15288631.55342549
>
> https://pasteboard.co/jI1eMKaj3PT1.png
>
> This is one in EPSG:3411:
>
> localhost:8000/wms?REQUEST=GetMap&SERVICE=WMS&VERSION=1.3.0&FORMAT=image%2Fpng&STYLES=wind_barb&TRANSPARENT=TRUE&TIME=2025-10-06T16%3A00%3A00Z&DIM_PRESSURE=50.0&DIM_REFERENCE_TIME=2025-10-06T03%3A00%3A00Z&LAYERS=wind_pl_vector_meps_det_vdiv_2_5km_calculations&WIDTH=1438&HEIGHT=1504&CRS=EPSG%3A3411&BBOX=404935.5094684181%2C-3769428.4764078585%2C4977748.7994076125%2C1013263.5876175575
>
> https://pasteboard.co/IHZjgXibtLyJ.png
>
> Thanks,
> Alistair
>
>
> On Sun, 5 Oct 2025 at 10:00, Seth G <sethg at geographika.co.uk> wrote:
>
>
> Hi Alistair,
>
> Could you give an example request (I presume it is WMS?)? Are you rotating
> your map at all, or is it just the projection?
>
> There are some tests which reproject a UVRASTER LAYER at
> https://github.com/MapServer/MapServer/blob/29e4ae81620a2c50201101c763df0b9d8d540b67/msautotest/wxs/wms_uvraster_map_reprojection.map
> It sounds like an oversight, so if you are able to create a standalone
> test case, it would be worth opening an issue.
>
> Seth
>
> --
> web:https://geographika.net & https://mapserverstudio.net
> mastodon: @geographika at mastodon.social
>
> On Fri, Oct 3, 2025, at 11:59 AM, Alistair Everett via MapServer-users
> wrote:
>
> Hi all,
>
> I'm trying to use the UVRASTER connection to plot wind barbs. I noticed if
> I request the layer in a different projection the location of the arrows
> are reprojected but the angle, and presumably vector components, are not -
> rather they remain in the same direction relative to my screen. It's a
> little unclear from the documentation if this is the expected behaviour for
> UVRASTER or not? I included a pasteboard link to a screenshot to show what
> I mean, compare for example vectors in the corners of the data and you will
> see they point in the same direction relative to my screen, rather than
> being rotated to match the projection.
>
> If this is expected, does anyone have any suggested ways of working around
> this?
>
> If this is not expected, does anyone have any ideas what I've done wrong?
> I can provide some sample data if someone wants to try to reproduce.
>
> I'm using: MapServer version 8.4.0 PROJ version 9.4 GDAL version 3.11
> OUTPUT=PNG OUTPUT=JPEG SUPPORTS=PROJ SUPPORTS=AGG SUPPORTS=FREETYPE
> SUPPORTS=CAIRO SUPPORTS=ICONV SUPPORTS=FRIBIDI SUPPORTS=WMS_SERVER
> SUPPORTS=WMS_CLIENT SUPPORTS=WFS_SERVER SUPPORTS=WFS_CLIENT
> SUPPORTS=WCS_SERVER SUPPORTS=SOS_SERVER SUPPORTS=OGCAPI_SERVER
> SUPPORTS=FASTCGI SUPPORTS=THREADS SUPPORTS=GEOS SUPPORTS=PBF INPUT=JPEG
> INPUT=OGR INPUT=GDAL INPUT=SHAPEFILE INPUT=FLATGEOBUF
>
> Screenshot: https://pasteboard.co/WTgbexDmQaAe.png
>
> Example mapfile: https://pastebin.com/ibPc6gVK
>
> Thanks,
> Alistair
>
> -----------------------------
> Dr Alistair Everett
> Senioringeniør / Senior Software Engineer
> Avdeling for Geoutvikling
>
> Phone: +47 939 68 985
> Email: alistair.everett at met.no
>
> Meteorologisk Institutt,
> Henrik Mohns plass 1,
> 0313 Oslo
> -----------------------------
> *Dr Alistair Everett*
> Senioringeniør / Senior Software Engineer
> Avdeling for Geoutvikling
>
> *Phone:* +47 939 68 985
> *Email:* alistair.everett at met.no <alistaire at met.no>
>
> *Meteorologisk Institutt*,
> Henrik Mohns plass 1,
> 0313 Oslo
>
> _______________________________________________
> MapServer-users mailing list
> MapServer-users at lists.osgeo.org
> https://lists.osgeo.org/mailman/listinfo/mapserver-users
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/mapserver-users/attachments/20251014/b004b863/attachment.htm>